Connect with us


MWC: AI meets IoT for AIoT in the real world



Until recently, artificial intelligence (AI) and the Internet of Things (IoT) were buzzwords for selling updated versions of existing communications and technology products. But now, the use cases are coming thick and fast, especially as the two technologies combine into a unified technology called AIoT.

At Mobile World Congress in Barcelona this week, SAS Institute is demonstrating how AIoT can help everyday virtual assistants through artificial emotional intelligence and use image-based intelligence for public safety, among other.

During MWC 2019, SAS is showcasing the following use cases for advanced analytics and AI applications:

  • Public Safety Monitoring powered by 5G and AI Computer Vision

The continued threat of terrorism all around the world has, in recent years, seen a change in tactics with more attacks performed utilising vehicle rampage, or mass gatherings of armed extremists targeting unsuspecting civilians going about their normal activities in our cities. AI computer vision techniques can be used to analyse potential threats from streaming video sources, and will increasing play an important role in keeping our cities and historical tourist attractions safer. Telco operators play a vital role in the AIoT ecosystems that provide these capabilities, where analysing data and responding “in the moment” could make the difference.

  • Real time monitoring and AI/ML anomalies detection for Network Service Quality

With the upcoming virtual network functions (NFV) and virtualization of Telco networks, real time monitoring and anomalies detection plays an important role without which “closed loop monitoring” system are impossible to function. Finding the “pre failing signatures”, meaning a combination of network KPI’s which lead to service degradation, or complete stop, among thousands on network KPI’s is impossible without the use of machine learning algorithms and deep learning.

  • Artificial Emotional Intelligence

After digital transformation programs, interactions between companies and their customers have been virtualized. They have lost knowledge about their customers. As human communications are based on emotional processes, automatic systems should adopt the proper identity to establish closer and lasting links. Thanks to AI, the same request sent to a virtual assistant is customized based on different users psychological profile.

  • Proactive Fraud Detection using AI

Telecommunications fraud is skyrocketing globally with over 30 billion dollars of losses reported last year by the CFCA. The biggest contributor to these fraud losses is Subscription and Dealer Fraud which typically account for 5-10% of annual revenues.

By harnessing the power of data analytics and machine learning, telecom operators can significantly reduce their fraud losses, optimize their operational costs and deliver a better, more secure customer experience.